Two key features are of note:
| Name | Newspapers |
| Number of rows | 61 |
| Number of columns | 7 |
| _______________________ | |
| Column type frequency: | |
| factor | 1 |
| numeric | 3 |
| POSIXct | 3 |
| ________________________ | |
| Group variables | None |
Variable type: factor
| skim_variable | n_missing | complete_rate | ordered | n_unique | top_counts |
|---|---|---|---|---|---|
| Location | 0 | 1 | FALSE | 2 | Hen: 36, Tia: 25 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 | hist |
|---|---|---|---|---|---|---|---|---|---|---|
| Price | 0 | 1.00 | 3.88 | 0.15 | 3.52 | 3.77 | 3.92 | 4.00 | 4.07 | ▂▃▃▅▇ |
| HHPrice | 25 | 0.59 | 3.80 | 0.14 | 3.52 | 3.69 | 3.80 | 3.90 | 4.07 | ▃▆▇▆▃ |
| TianjinPrice | 36 | 0.41 | 3.99 | 0.05 | 3.88 | 3.98 | 4.00 | 4.04 | 4.05 | ▂▂▂▇▇ |
Variable type: POSIXct
| skim_variable | n_missing | complete_rate | min | max | median | n_unique |
|---|---|---|---|---|---|---|
| Date | 0 | 1.00 | 2013-05-04 | 2013-10-30 | 2013-08-06 00:00:00 | 61 |
| HHDate | 25 | 0.59 | 2013-05-04 | 2013-10-25 | 2013-07-29 12:00:00 | 36 |
| TianjinDate | 36 | 0.41 | 2013-05-05 | 2013-10-30 | 2013-08-13 00:00:00 | 25 |
Two key features are of note:
Almost everything that we wish to know can be discerned from this.
The Raw Table
Location Build No Total
HenryHub 66 15 81
Tianjin 31 33 64
Total 97 48 145
A Percentage Table
Location Build No
HenryHub 0.8148148 0.1851852
Tianjin 0.4843750 0.5156250
Two things are of note:
1. The data are paired by period; that is the point of the calibration exercise.
2. Henry Hub is almost always forecast higher than Tianjin. Choose a period and you can hover over the associated values to convince yourself of this. Compare the two forecasts for any given period; they are almost always higher for Henry Hub.
Henry Hub is almost always forecast higher than Tianjin. The difference column captures this perfectly and shows that it always favors Henry Hub.